Conversational Case Base Recommender Systems for Metadata Discovery Mehmet S. Aktas, Marlon Pierce, Geoffrey Fox and David Leake Indiana University.

Slides:



Advertisements
Similar presentations
Dr. Leo Obrst MITRE Information Semantics Information Discovery & Understanding Command & Control Center February 6, 2014February 6, 2014February 6, 2014.
Advertisements

School of something FACULTY OF OTHER School of Computing FACULTY OF ENGINEERING Formalising a basic hydro-ontology David Mallenby Knowledge Representation.
CH-4 Ontologies, Querying and Data Integration. Introduction to RDF(S) RDF stands for Resource Description Framework. RDF is a standard for describing.
So What Does it All Mean? Geospatial Semantics and Ontologies Dr Kristin Stock.
Semantic Web & Cased Based Reasoning AIST Meeting JPL, CA 2003 Mehmet S. Aktas
E-DECIDER Workshop: QuakeSim Tools and Products Marlon Pierce Indiana University.
Research topics Semantic Web - Spring 2007 Computer Engineering Department Sharif University of Technology.
Distributed components
Information Retrieval in Practice
Semantic Web and Web Mining: Networking with Industry and Academia İsmail Hakkı Toroslu IST EVENT 2006.
Descriptions Robert Grimm New York University. The Final Assignment…  Your own application  Discussion board  Think: Paper summaries  Web cam proxy.
Adaptive Hypermedia Meets Provenance Evgeny Knutov Paul De Bra Mykola Pechenizkiy GAF project: Generic Adaptation Framework (project is supported byNWO.
Automatic Data Ramon Lawrence University of Manitoba
MCE 561 Computational Methods in Solid Mechanics
Overview of Search Engines
Web-based Portal for Discovery, Retrieval and Visualization of Earth Science Datasets in Grid Environment Zhenping (Jane) Liu.
Introduction to virtual engineering László Horváth Budapest Tech John von Neumann Faculty of Informatics Institute of Intelligent Engineering.
Becerra-Fernandez, et al. -- Knowledge Management 1/e -- © 2004 Prentice Hall Chapter 16 Knowledge Application Systems: Systems that Utilize Knowledge.
Help!!! Some Future Semantic Grid Activities CrisisGrid and ServoGrid PTLIU Laboratory for Community Grids Geoffrey Fox Computer Science, Informatics,
BiodiversityWorld GRID Workshop NeSC, Edinburgh – 30 June and 1 July 2005 Metadata Agents and Semantic Mediation Mikhaila Burgess Cardiff University.
1 Yolanda Gil Information Sciences InstituteJanuary 10, 2010 Requirements for caBIG Infrastructure to Support Semantic Workflows Yolanda.
EARTH SCIENCE MARKUP LANGUAGE “Define Once Use Anywhere” INFORMATION TECHNOLOGY AND SYSTEMS CENTER UNIVERSITY OF ALABAMA IN HUNTSVILLE.
CS621 : Seminar-2008 DEEP WEB Shubhangi Agrawal ( )‏ Jayalekshmy S. Nair ( )‏
Geoffrey Fox Andrea Donnellan May 3, 2004 Network and Grid Computing Computational Geoinformatics Workshop.
Chapter 7 Web Content Mining Xxxxxx. Introduction Web-content mining techniques are used to discover useful information from content on the web – textual.
Clément Troprès - Damien Coppéré1 Semantic Web Based on: -The semantic web -Ontologies Come of Age.
ISERVO International Solid Earth Research Virtual Observatory Grid/Web Services and Portals Supporting Earthquake Science December AGU Fall Meeting.
Publishing and Visualizing Large-Scale Semantically-enabled Earth Science Resources on the Web Benno Lee 1 Sumit Purohit 2
Deploying Trust Policies on the Semantic Web Brian Matthews and Theo Dimitrakos.
LIS 506 (Fall 2006) LIS 506 Information Technology Week 11: Digital Libraries & Institutional Repositories.
The MMI Tools Carlos Rueda Monterey Bay Aquarium Research Institute OOS Semantic Interoperability Workshop Marine Metadata Interoperability Project Boulder,
RDF and OWL Developing Semantic Web Services by H. Peter Alesso and Craig F. Smith CMPT 455/826 - Week 6, Day Sept-Dec 2009 – w6d21.
School of Computing FACULTY OF ENGINEERING Developing a methodology for building small scale domain ontologies: HISO case study Ilaria Corda PhD student.
Scalable Metadata Definition Frameworks Raymond Plante NCSA/NVO Toward an International Virtual Observatory How do we encourage a smooth evolution of metadata.
DDI-RDF Leveraging the DDI Model for the Linked Data Web.
EARTH SCIENCE MARKUP LANGUAGE Why do you need it? How can it help you? INFORMATION TECHNOLOGY AND SYSTEMS CENTER UNIVERSITY OF ALABAMA IN HUNTSVILLE.
EU Project proposal. Andrei S. Lopatenko 1 EU Project Proposal CERIF-SW Andrei S. Lopatenko Vienna University of Technology
Metadata. Generally speaking, metadata are data and information that describe and model data and information For example, a database schema is the metadata.
GEM Portal and SERVOGrid for Earthquake Science PTLIU Laboratory for Community Grids Geoffrey Fox, Marlon Pierce Computer Science, Informatics, Physics.
Semantic Web - an introduction By Daniel Wu (danielwujr)
Knowledge Representation of Statistic Domain For CBR Application Supervisor : Dr. Aslina Saad Dr. Mashitoh Hashim PM Dr. Nor Hasbiah Ubaidullah.
2007. Software Engineering Laboratory, School of Computer Science S E Web-Harvest Web-Harvest: Open Source Web Data Extraction tool 이재정 Software Engineering.
Lifecycle Metadata for Digital Objects November 1, 2004 Descriptive Metadata: “Modeling the World”
Current and Potential Uses for GIS in Academic Arctic Research Michael F. Goodchild University of California Santa Barbara.
EEL 5937 Ontologies EEL 5937 Multi Agent Systems Lecture 5, Jan 23 th, 2003 Lotzi Bölöni.
Semantic Technologies and Application to Climate Data M. Benno Blumenthal IRI/Columbia University CDW /04-01.
GEON2 and OpenEarth Framework (OEF) Bradley Wallet School of Geology and Geophysics, University of Oklahoma
ISERVO and SERVOGrid: (International) Solid Earth Research Virtual Observatory Grid/Web Services and Portals Supporting Earthquake Science Jan
OWL Representing Information Using the Web Ontology Language.
Metadata Common Vocabulary a journey from a glossary to an ontology of statistical metadata, and back Sérgio Bacelar
Of 33 lecture 1: introduction. of 33 the semantic web vision today’s web (1) web content – for human consumption (no structural information) people search.
Shridhar Bhalerao CMSC 601 Finding Implicit Relations in the Semantic Web.
Distributed Data Analysis & Dissemination System (D-DADS ) Special Interest Group on Data Integration June 2000.
Digital Video Library Network Supervisor: Prof. Michael Lyu Student: Ma Chak Kei, Jacky.
Supported in part by the National Science Foundation under Grant No. HRD Any opinions, findings, and conclusions or recommendations expressed.
Semantic Data Extraction for B2B Integration Syntactic-to-Semantic Middleware Bruno Silva 1, Jorge Cardoso 2 1 2
Application Web Service Toolkit Allow users to quickly add new applications GGF5 Edinburgh Geoffrey Fox, Marlon Pierce, Ozgur Balsoy Indiana University.
CIMA and Semantic Interoperability for Networked Instruments and Sensors Donald F. (Rick) McMullen Pervasive Technology Labs at Indiana University
Selected Semantic Web UMBC CoBrA – Context Broker Architecture  Using OWL to define ontologies for context modeling and reasoning  Taking.
Interacting Data Services for Distributed Earthquake Modeling Marlon Pierce, Choonhan Youn, and Geoffrey Fox Community Grids Lab Indiana University.
Metayogi Increasing the Accessibility of the Semantic Web Karim Tharani Doug Macdonald Rachel Heidecker.
Semantic metadata in the Catalogue Frédéric Houbie.
1 Implementing Geographic Information System Grid Services Using Distributed Messaging Systems Marlon Pierce Community Grids Lab Indiana University December.
Information Retrieval in Practice
SERVO Grid: Solid Earth Research Virtual Observatory Grid/Web Services and Portals Supporting Earthquake Science July Fourth ACES APEC Cooperation.
Ontology.
GeoFEST tutorial What is GeoFEST?
OVERVIEW OF FINITE ELEMENT METHOD
HAO/SCD: VO, metadata, catalogs, ontologies, querying
Web archives as a research subject
Presentation transcript:

Conversational Case Base Recommender Systems for Metadata Discovery Mehmet S. Aktas, Marlon Pierce, Geoffrey Fox and David Leake Indiana University

Solid Earth Research Virtual Observatory Grid (SERVOGrid) SERVOGrid is a NASA project to integrate historical, measured, and calculated earthquake data (GPS, Seismicity, Fault) with simulation codes. SERVOGrid is a NASA project to integrate historical, measured, and calculated earthquake data (GPS, Seismicity, Fault) with simulation codes. SERVOGrid resources located at various institutions across the country. SERVOGrid resources located at various institutions across the country. # of resources, services and their usage frequency expected to grow quickly. # of resources, services and their usage frequency expected to grow quickly.

Characteristics of Computing for Solid Earth Science Widely distributed datasets in various formats Widely distributed datasets in various formats GPS, Fault data, Seismic data sets, InSAR satellite dataGPS, Fault data, Seismic data sets, InSAR satellite data Many available in state of art tar files that can be FTP’dMany available in state of art tar files that can be FTP’d Distributed models and expertise Distributed models and expertise Lots of codes with different regions of validity, ranging from cellular automata to finite element to data mining applications (HMM)Lots of codes with different regions of validity, ranging from cellular automata to finite element to data mining applications (HMM) Some codes also have export or IP restrictionsSome codes also have export or IP restrictions Other codes are highly specialized to their deployment environments.Other codes are highly specialized to their deployment environments. Decomposable problems requiring interoperability for linking full models Decomposable problems requiring interoperability for linking full models The fidelity of your fault modeling can vary considerablyThe fidelity of your fault modeling can vary considerably Link codes (through data) to support multiple scalesLink codes (through data) to support multiple scales

SERVOGrid Applications Codes range from simple “rough estimate” codes to parallel, high performance applications. Codes range from simple “rough estimate” codes to parallel, high performance applications. Disloc: handles multiple arbitrarily dipping dislocations (faults) in an elastic half-space.Disloc: handles multiple arbitrarily dipping dislocations (faults) in an elastic half-space. Simplex: inverts surface geodetic displacements for fault parameters using simulated annealing downhill residual minimization.Simplex: inverts surface geodetic displacements for fault parameters using simulated annealing downhill residual minimization. GeoFEST: Three-dimensional viscoelastic finite element model for calculating nodal displacements and tractions. Allows for realistic fault geometry and characteristics, material properties, and body forces.GeoFEST: Three-dimensional viscoelastic finite element model for calculating nodal displacements and tractions. Allows for realistic fault geometry and characteristics, material properties, and body forces. Virtual California: Program to simulate interactions between vertical strike-slip faults using an elastic layer over a viscoelastic half-spaceVirtual California: Program to simulate interactions between vertical strike-slip faults using an elastic layer over a viscoelastic half-space RDAHMM: Time series analysis program based on Hidden Markov Modeling. Produces feature vectors and probabilities for transitioning from one class to another.RDAHMM: Time series analysis program based on Hidden Markov Modeling. Produces feature vectors and probabilities for transitioning from one class to another. Preprocessors, mesh generators: AKIRA suite Preprocessors, mesh generators: AKIRA suite Visualization tools: RIVA, GMT, IDL Visualization tools: RIVA, GMT, IDL

Motivation Most fundamental challenge is just making these codes useable for other researchers. Most fundamental challenge is just making these codes useable for other researchers. And hooking these codes to data sources And hooking these codes to data sources First step is to express resources with descriptive metadata First step is to express resources with descriptive metadata Then explore intelligent retrieval mechanisms to make these resources available Then explore intelligent retrieval mechanisms to make these resources available

SERVOGrid Ontology Overview We have a collection of codes, visualization tools, computing resources, and data sets that we want to combine in an ontology. We have a collection of codes, visualization tools, computing resources, and data sets that we want to combine in an ontology. Ontology instances can then be built to describe specific resources. Ontology instances can then be built to describe specific resources. After we have built instances, we can pose queries on the data to retrieve values. After we have built instances, we can pose queries on the data to retrieve values. Values may be structured, so we can do “stepped” queries.Values may be structured, so we can do “stepped” queries. We thus need to start by grouping together related resources. We thus need to start by grouping together related resources.

An Instance for Disloc code <rdf:RDF xmlns:rdf=' xmlns:rdfs=' xmlns:rdfs=' xmlns:servo=' xmlns:servo=' xmlns:dc=" xmlns:dc=" A. Donnellan A. Donnellan </rdf:Description></rdf:RDF>

From SW Representation to Case Base Reasoning (CBR) Developing new tools, applications and architectures on top of the Semantic Web is the real challenge. Developing new tools, applications and architectures on top of the Semantic Web is the real challenge. Can we ensure consistency and correctness in the presentation of information??Can we ensure consistency and correctness in the presentation of information?? AI techniques could be considered as basis for a resource recommender system. AI techniques could be considered as basis for a resource recommender system. CBR is most suitable AI technique for SERVOGrid domain. CBR is most suitable AI technique for SERVOGrid domain.

What is Case-Based Reasoning? (CBR in a Nutshell) CBR is reasoning by remembering CBR is reasoning by remembering In CBR, recommendations are made by doing reasoning from current set of cases In CBR, recommendations are made by doing reasoning from current set of cases Classification CBR Classification CBR when a similar problem description is entered most similar cases are suggested (by comparing and contrasting problem description with current set of cases) to the user as resultswhen a similar problem description is entered most similar cases are suggested (by comparing and contrasting problem description with current set of cases) to the user as results

Conversational CBR (CCBR) (CCBR in a Nutshell) CCBR is a type of CBR that relies on question- answer sessions to recommend most similar cases. CCBR is a type of CBR that relies on question- answer sessions to recommend most similar cases. User interacts with the system to fill in the gaps to retrieve right cases User interacts with the system to fill in the gaps to retrieve right cases System responds with ranked cases and questions at each step System responds with ranked cases and questions at each step Question-answer-ranking cycle continues until success or failure Question-answer-ranking cycle continues until success or failure success: if user finds an answer to his querysuccess: if user finds an answer to his query failure: if no satisfactory case is foundfailure: if no satisfactory case is found

What is a Case? (CCBR Case in SERVOGrid) A case is composed of: A case is composed of: problem description: metadata concerning desired characteristics of a SERVOGrid resource, e.g., RDF triples describing a resourceproblem description: metadata concerning desired characteristics of a SERVOGrid resource, e.g., RDF triples describing a resource solution: pointer to a resource described by metadata in problem descriptionsolution: pointer to a resource described by metadata in problem description A Casebase is library of cases generated from file store of RDF files each representing a case. A Casebase is library of cases generated from file store of RDF files each representing a case.

CCBR CASE Problem Solution RDF Triple =(Subject, Predicate, Object) CCBR Case with RDF Representation RDF Triple

CCBR Recommender System Ranking of the cases Ranking of the cases Cases will be ranked based on their consistent triple numbers.Cases will be ranked based on their consistent triple numbers. If the case has a matching triple, it will have higher ranking.If the case has a matching triple, it will have higher ranking. If the case does not have the entered triple, its ranking won’t change, unless user wants the cases which don’t have this triple.If the case does not have the entered triple, its ranking won’t change, unless user wants the cases which don’t have this triple. Ranking of the questions Ranking of the questions Ranking can be based on (property, property value) appearance # in the triples stored in the case base.Ranking can be based on (property, property value) appearance # in the triples stored in the case base. System must recommend good starting points for user specification of servoObject class properties.System must recommend good starting points for user specification of servoObject class properties.

CCBR CASEBASE Case Feature 1 Feature 2 Feature 5 Feature 1 Feature 2 Feature 5 Case = Feature 1 Feature 2 Feature 3 Feature 4 Feature 1 Feature 2 Feature 3 Feature 4 A Case from CASEBASE Query Case IF ((A.Feature1.Solution = B.Feature1.Solution) & (A.Feature2.Solution = B.Feature2.Solution)) THEN Consistency # = 2 AB CCBR Recommender System

Recap: SERVOGrid Case Base Recommender System goal: locating resources in a large scale environment (SERVOGrid project) goal: locating resources in a large scale environment (SERVOGrid project) approach: approach: SERVOGrid ontology instances (metadata) to describe resourcesSERVOGrid ontology instances (metadata) to describe resources Recommender system to aid metadata discoveryRecommender system to aid metadata discovery Conversational CBR with SW markup languages providing standard form for case representationConversational CBR with SW markup languages providing standard form for case representation

More Information SERVOGrid/QuakeSim: SERVOGrid/QuakeSim: SERVOGrid Recommender Systems project: SERVOGrid Recommender Systems project: htmlhttp://tambora.ucs.indiana.edu/~maktas/servo/project. htmlhttp://tambora.ucs.indiana.edu/~maktas/servo/project. htmlhttp://tambora.ucs.indiana.edu/~maktas/servo/project. html SERVOGrid Recommender Systems demo: SERVOGrid Recommender Systems demo: sphttp://ripvanwinkle.ucs.indiana.edu:4780/cbr/selection.j sp Publications: Publications:

Questions/Comments Any questions and/or comments? Any questions and/or comments? Thanks! Thanks!